Home/Directory/AI & ML/AI Vision

AI Vision

@tan-yong-sheng · github.com/tan-yong-sheng/ai-vision-mcp

Multimodal AI vision MCP server for image, video, and object detection analysis. Enables UI/UX evaluation, visual regression testing, and interface understanding using Google Gemini and Vertex AI.

ClaimedAI & MLstdioToolslocal auth
GitHub

Install one command

terminal
$ npx ai-vision-mcp

Or add it with the Claude CLI: claude mcp add tan-yong-sheng-ai-vision-mcp

Tools exposed 4 tools

analyze_image
compare_images
detect_objects_in_image
audit_design

Configuration generate & copy

Values are filled in locally in your browser — nothing is sent anywhere.

claude_desktop_config.json
{
  "mcpServers": {
    "tan-yong-sheng-ai-vision-mcp": {
      "command": "npx",
      "args": ["ai-vision-mcp"],
      "env": {
        "GEMINI_API_KEY": "<your-gemini-api-key>",
        "MAX_TOKENS_FOR_DETECT_OBJECTS_IN_IMAGE": "<your-max-tokens-for-detect-objects-in-image>",
        "MAX_TOKENS": "<your-max-tokens>"
      }
    }
  }
}

Paste the Claude Desktop snippet under mcpServers in your config, then restart Claude Desktop — or run the Claude Code command in your terminal.

Related

Similar servers

More in AI & ML
Compare
@microsoft

MCP tool access to MarkItDown a library that converts many file formats (local or remote) to Markdown for LLM consumption.

AI & MLstdioRead-only no-auth
149.5k3w ago5 toolslocal
$pip install -e 'packages/markitdown[all]'
Compare
@upstash

Up-to-date code documentation for LLMs and AI code editors.

AI & MLstdioRead-only auth
57.1k6d ago3 toolslocal
$npx ctx7 setup
Compare
@eyaltoledano

AI-powered task management system for AI-driven development.

AI & MLstdioWrite auth
27.4k2mo ago15 toolslocal
$npx task-master init